基于部分平行ADMM求解谱寻求问题的时频分析

doi:10.3969/j.issn.0372-2112.2019.11.022

电子学报 ›› 2019, Vol. 47 ›› Issue (11): 2392-2398.DOI: 10.3969/j.issn.0372-2112.2019.11.022

基于部分平行ADMM求解谱寻求问题的时频分析

胡园园, 罗倩, 段中钰, 王嘉浩

北京信息科技大学信息与通信工程学院, 北京 100020

收稿日期:2018-12-17 修回日期:2019-05-13 出版日期:2019-11-25
- 通讯作者:
- 胡园园
- 作者简介:
- 罗倩女,1965年生于山西太原,现为北京信息科技大学副教授,硕士生导师.主要研究方向为信号与信息处理,大数据处理.E-mail:luoqian@bistu.edu.com;段中钰 男,1977年生于黑龙江鹤岗,现为北京信息科技大学副教授,硕士生导师.主要研究方向为信号与信息处理,数据挖掘.E-mail:duanzhongyu@bistu.edu.cn
- 基金资助:
- 北京市教委科研计划项目 (No.KM201811232009)

Time-Frequency Analysis Based on Partly Parallel ADMM Solving Spectral Pursuit Problem

HU Yuan-yuan, LUO Qian, DUAN Zhong-yu, WANG Jia-hao

School of Information and Communication Engineering, Beijing Information Science&Technology University, Beijing 100020, China

Received:2018-12-17 Revised:2019-05-13 Online:2019-11-25 Published:2019-11-25
- Supported by:
- Research Program of Beijing Municipal Education Commission (No.KM201811232009)

摘要/Abstract

摘要： 经典的非参数谱分析方法使用滑动窗口来捕捉大多数时间序列的频谱特性，然而这种方法不能很好地应用在时间序列的时频谱是时间连续的信号上.对于一些其时频谱满足时间连续频率稀疏的非平稳信号，提出了一种利用部分平行交替方向乘子法（Alternating Direction Method of Multipliers，ADMM）求解谱寻求问题用于此类信号的时频分析方法.对一段加噪声的仿真信号和一段EEG（脑电）信号使用提出的方法进行时频分析.仿真结果表明：与短时傅里叶的分析方法相比，提出的方法不仅提高了时频谱的频率分辨率和时间分辨率，还有效抑制了噪声.最后从ADMM算法停止准则的角度说明了算法的收敛.

关键词: 非参数谱分析, 滑动窗口, 部分平行ADMM, 谱寻求, 时频分析, 频率分辨率, 时间分辨率

Abstract: Classical nonparametric spectral analysis uses sliding windows to capture the dynamic nature of most real-world time series,however,this universally accepted approach fails to exploit the temporal continuity in the data.For some non-stationary signals that are smooth in time and sparse in frequency,a method of computing the spectrum pursuit estimate by using partly parallel alternating direction method of multipliers (ADMM) was used to obtain time-frequency analysis of this signal.A time-frequency analysis was performed on simulated and real human EEG data using the proposed method.The simulation results show that the proposed method not only effectively improves time and frequency resolution of the spectrum,but also effectively suppresses the noise than short-time Fourier transform.Finally,the convergence of the algorithm was illustrated from the perspective of stopping criterion of ADMM algorithm.

Key words: nonparametric spectral analysis, sliding windows, partly parallel Alternating Direction Method of Multipliers (ADMM), spectrum pursuit, time-frequency analysis, frequency resolution, time resolution

中图分类号:

TN911.6

胡园园, 罗倩, 段中钰, 等. 基于部分平行ADMM求解谱寻求问题的时频分析[J]. 电子学报, 2019, 47(11): 2392-2398.

HU Yuan-yuan, LUO Qian, DUAN Zhong-yu, et al. Time-Frequency Analysis Based on Partly Parallel ADMM Solving Spectral Pursuit Problem[J]. Acta Electronica Sinica, 2019, 47(11): 2392-2398.

参考文献

[1] 张贤达.现代信号处理[M].北京:清华大学出版社,2015.286-287. Zhang Xian-da.Modern Signal Processing[M].Beijing:Tsinghua University Press,2015.286-287.(in Chinese)
[2] Purdon PL,Pierce ET,Mukamel EA,et al.Electroencephalogram signatures of loss and recovery of consciousness from propofol[J].Proceedings of the National Academy of Sciences,2013,110(2):E1142-E1151.
[3] Wu Z,Huang NE.Ensemble empirical mode decomposition:A noise-assisted data analysis method[J].Advances in Adaptive Data Analysis,2009,1(1):1-41.
[4] Ba D,Babadi B,Purdon PL.Robust spectrotemporal decomposition by iteratively reweighted least squares[J].Proceedings of the National Academy of Sciences,2014,111(50):E5336-E5345.
[5] Ziniel J,Potter LC,Schniter P.Tracking and smoothing of time-varying sparse signals via approximate belief propagation[J].Signals,Systems & Computers,2010,45(2):808-812.
[6] Adam Charles,Aurele Balavoine,Christopher Rozell.Dynamic filtering of time-varying sparse signals via l1 minimization[J].IEEE Transactions on Signal Processing,2015,64(21):5644-5656.
[7] Charles A,Asif M S,Romberg J.Sparsity penalties in dynamical system estimation[A].Barbara A Sullivan.2011 45th Annual Conference on Information Sciences and Systems[C].Baltimore,MD,USA:IEEE,2011.1-6.
[8] Hastie T,Tibshirani R,Wainwright M.Statistical Learning with Sparsity[M].Boca Raton:CRC press,2015.
[9] Schamberg Gabriel,Ba Demba,Wagner Mark.Efficient low-rank spectrotemporal decomposition using ADMM[A].2016 IEEE Statistical Signal Processing Workshop (SSP)[C].Palma de Mallorca,Spain:IEEE,2016.1-5.
[10] Schamberg Gabriel,Ba Demba,Coleman Todd.A modularized efficient framework for non-markov time series estimation[J].IEEE Transactions on Signal Processing,2018,66(12):3140-3154.
[11] 周志华.机器学习[M].北京:清华大学出版社,2016.132-133. Zhou Zhi-hua.Machine Learning[M].Beijing:Tsinghua University Press,2016.132-133.(in Chinese)
[12] Aleksandr Y Aravkin,James V Burke.Compressed Sensing & Sparse Filtering[M].Berlin,Heidelberg:Springer,2013.237-280.
[13] Shiqian Ma,Donald Goldfarb,Lifeng Chen.Fixed point and Bregman iterative methods for matrix rank minimization[J].Mathematical Programming,2011,128(1-2):321-353.
[14] Tibshirani R,Saunders M,Rosset S,et al.Sparsity and smoothness via the fused lasso[J].Journal of the Royal Statistical Society,2005,67(1):91-108.
[15] 韩清鹏.利用EEG信号的小波包变换与非线性分析实现精神疲劳状态的判定[J].振动与冲击,2013,32(2):182-187. Han Qing-peng.Evaluation of human mental stress states based on wavelet package transformation and nonlinear analysis of EEG signals[J].Journal of Vibration and Shock,2013,32(2):182-187.(in Chinese)
[16] Kang DY,Kim YS,Ornelas G,et al.Scalable microfabrication procedures for adhesive-integrated flexible and stretchable electronic sensors[J].Sensors,2015,15(9):23459-23476.
[17] Boyd S,Parik N,Chu E,et al.Distributed optimization and statistical learning via the alternating direction method of multipliers[J].Foundations and Trends in Machine Learning,2011,3(1):1-122.

基于部分平行ADMM求解谱寻求问题的时频分析

Time-Frequency Analysis Based on Partly Parallel ADMM Solving Spectral Pursuit Problem

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	李钦, 刘伟, 牛朝阳, 宝音图, 惠周勃. 低信噪比下基于分裂EfficientNet网络的雷达信号调制方式识别[J]. 电子学报, 2023, 51(3): 675-686.
[2]	张聿远, 张立民, 闫文君. 基于深度多级残差网络的低信噪比下空频分组码识别方法[J]. 电子学报, 2022, 50(1): 79-88.
[3]	李军, 孙君, 余云河, 贾志豪. 基于滑动窗口的接入新方案[J]. 电子学报, 2021, 49(2): 300-305.
[4]	刘金平, 王杰, 唐朝晖, 贺俊宾, 谢永芳, 马天雨. 基于自适应滑窗递归稀疏主成分分析的工业过程故障监测[J]. 电子学报, 2020, 48(9): 1795-1803.
[5]	王宇, 陈晨, 张家琦, 彭丝雨, 高军峰. 任务提示的脑认知加工过程中神经振荡机制研究[J]. 电子学报, 2020, 48(5): 960-965.
[6]	秦鑫, 黄洁, 查雄, 骆丽萍, 胡德秀. 基于扩张残差网络的雷达辐射源信号识别[J]. 电子学报, 2020, 48(3): 456-462.
[7]	刘秋红, 李天昀, 王彬, 许漫坤, 邱钊洋. 基于离散小波分解和频率脊线分析的CPM信号符号速率盲估计[J]. 电子学报, 2020, 48(3): 470-477.
[8]	查雄, 许漫坤, 彭华, 秦鑫, 李天昀. 基于深度残差网络的特定协议信号识别[J]. 电子学报, 2019, 47(7): 1532-1537.
[9]	李淑雅, 施明, 张建秋. 时频分析的卡尔曼滤波匹配解调变换法[J]. 电子学报, 2018, 46(4): 769-774.
[10]	郑近德, 潘海洋, 戚晓利, 张兴权, 刘庆运. 基于改进经验小波变换的时频分析方法及其在滚动轴承故障诊断中的应用[J]. 电子学报, 2018, 46(2): 358-364.
[11]	陈小龙, 关键, 于晓涵, 何友. 雷达动目标短时稀疏分数阶傅里叶变换域检测方法[J]. 电子学报, 2017, 45(12): 3030-3036.
[12]	郑近德, 潘海洋, 程军圣. 非平稳信号分析的广义解析模态分解方法[J]. 电子学报, 2016, 44(6): 1458-1464.
[13]	郑近德, 程军圣, 曾鸣. 基于改进的局部特征尺度分解和归一化正交的时频分析方法[J]. 电子学报, 2015, 43(7): 1418-1424.
[14]	孙新江, 吴晓蓓, 黄成, 徐志良. 无线传感器网络中自修复密钥颁布机制中滑动窗口的动态性探讨[J]. 电子学报, 2015, 43(3): 447-453.
[15]	庞存锁, 刘磊, 单涛. 基于短时分数阶傅里叶变换的时频分析方法[J]. 电子学报, 2014, 42(2): 347-352.